Twitter user @dog_rates, also known as WeRateDogs, is a Twitter account that rates people's dogs with a humorous comment about the dog.
These ratings almost always have a denominator of 10. The numerators, also almost always greater than 10. Example: 11/10, 12/10, 13/10, etc.
WeRateDogs downloaded their Twitter archive and sent it to Udacity via email exclusively for the students to use in this project. This archive contains basic tweet data (tweet ID, timestamp, text, etc.) for all 5000+ of their tweets as they stood on August 1, 2017.
Udacity also provided a image prediction file, of the top 3 predictions of the dogs associated with the image of each tweet.
Additional data such as favourites and retweets information on each tweet was downloaded via Tweepy (twitter api) through a twitter developer acount that I had to make.
Example:
The first step was the extract the favourites and retweets using the Tweepy.
After, the 3 dataframes were explored to see if I can find any data quality or tidiness issues. Quality issues are issues with data itself such as missing data or incorrectly formatted etc. Tidiness issues are structural issues such as are the columns neccessary or can it be simplified etc.
Essentially,I'm trying to keep the data consistent and accurate enough to analyze.
After cleaning all the issues found (9 quality and 4 tidy), I was ready to analyze!
As the project overview said, almost all dogs were given a higher rating than 10. We can see that most dogs got a rating of 10 to 13 with 14 and 15 being very rare and reserved for the best doggos and puppers. The graph shows the ratings to be left skewed.
Most tweets try to classify the dog into 1 of 4 stages: Pupper, Puppo, Doggo and Floofer (please see below pic for definitions). However, we can see that most tweets (84%) do not have a dog stage classification. Puppers have 10% of the tweets, little ones leading the pack, while doggos follow with 4% of tweets.
Top 10 dog breeds that are tweeted are really interesting. The golden retriever being number 1 with over 150 tweets. This can be expected as it is a common friendly dog breed and everyone loves a goldie. Followed by labrador retriever and pembroke in 2nd and 3rd.
This is a correlation between favourites and retweets. There is a positive correlation and it can be expected. The more retweets a picture gets, the more audience it reaches and the more favourites and additional retweets it can recieve.
One last post, featuring the cutest little pupper:
Thank you. Remember all dogs are good dogs.